In a new hospital ranking, doing good counts nearly as much as doing well

You won’t find the usual suspects like Massachusetts General Hospital or the Mayo Clinic at the top of a new ranking of U.S. hospitals. That’s because the rating system relies not just on traditional quality measures, but also on a hospital’s community-minded policies and avoidance of unnecessary care.

The rankings show that those hospitals with good clinical outcomes tend to score poorly in addressing inequities that affect the health of their communities. And even when hospitals perform many low-value procedures — those that research has shown to be of limited or no benefit — their patients generally have a low risk of dying both in the hospital and after being discharged, as well as a low risk of having to be readmitted.

No previous hospital rankings use “civic leadership,” which includes community-minded policies such as charity care, financial aid, and paying all staffers a living wage, or “value of care,” meaning whether a hospital avoids 13 procedures of questionable or clearly absent clinical benefit. The Lown Institute, a nonprofit think tank in Brookline, Mass., incorporated both measures into its rankings of 3,282 hospitals because “it is time for hospitals to rethink what it means to be great,” said Lown’s president, physician Vikas Saini.

Civic leadership matters because the health of the people in a hospital’s community “reflect things outside the four walls of a hospital,” he said. “If one patient is going back to a community without a lot of resources, where health equity is low, and another is going back to a wealthier community, their [long-term] health will be very different even if the quality of care in the hospital was identical. Hospitals have an obligation to improve the health of the outside community.”

Veterans of hospital rankings generally applauded Lown’s efforts, with caveats.

“I love that they’re measuring civic leadership,” said Sara Singer of Stanford University School of Medicine, an expert in health care management. “I also think they’re right to evaluate overuse of low-value procedures: You want to be sure you aren’t going to receive treatments you shouldn’t be receiving.”

She questioned, however, whether those measures should be bundled into a composite score, as Lown did. Quality of care indicators such as staffing levels and patient outcomes such as surgical mistakes and bedsores account for 50% of a hospital’s composite ranking and associated letter grade, while civic leadership counts for 30% and value of care 20%.

“I’m not sure civic leadership should influence where a patient chooses to get care as much as quality measures should,” Singer said, “though I can see using it to decide where you might make a charitable contribution.”

In contrast, patients would do well to consider a hospital’s overuse of procedures that research has shown to have little clinical benefit, she said, such as knee arthroscopy, spinal fusions, EEG for headache, hysterectomies for benign conditions such as uterine fibroids or pelvic organ prolapse, and putting stents in renal arteries.

Although a patient having, say, a hip replaced might not care whether a hospital performs many vertebroplasties (injecting bone “cement” into vertebrae to treat osteoporosis) or other low-value procedures unrelated to her planned surgery, Singer said, “I can imagine that if a hospital is overtreating patients with these particular procedures, it might overmedicate, oversedate, or overtreat you for what you’re there for, too.” In addition, some low-value care disproportionately affects people of color: Black women are more likely to have unnecessary hysterectomies than white women are.

Many hospitals perform high numbers of low-quality procedures, which tend to be lucrative (and waste an estimated $100 billion per year in health care spending). “The right thing to do financially isn’t necessarily the right thing to do for patients,” said Leah Binder, president and CEO of the Leapfrog Group, a nonprofit backed by large employers that aims to identify and promote high-quality, high-value health care and whose hospital ratings emphasize patient safety.

She called Lown’s decision to downgrade a hospital for performing many low-quality procedures “a breakthrough” in hospital ratings. “We know inappropriate care is an extremely significant problem for the health care system,” Binder said. “It’s also a terrible tragedy for individual patients to go through an unnecessary procedure.”

The American Hospital Association, however, said in a statement that the report offers consumers no “accurate and useful information,” instead offering “a hodgepodge of composite score, ranking, star ratings, and letter grades that will, at best, confuse consumers and likely mislead them.”

In Lown’s composite rankings, which are based on 2017 data from Medicare and other government sources, the top hospitals are JPS Health Network in Fort Worth, Texas; Marshall Medical Center in Placerville, Calif.; UPMC McKeesport in Pennsylvania; Seton Northwest Hospital in Austin, Texas; and Mercy Health-West Hospital in Cincinnati. Each scored A or A+ on all three measures, showing it is possible to do well by the community while also delivering excellent patient care, Saini said.

Of course, hospital rankings reflect the values of the groups that produce them. For instance, JPS gets an underwhelming C  from Leapfrog, due to safety failures (patient bedsores and falls), unresponsive doctors and nurses, and surgical problems (wounds splitting open and dangerous blood clots).

Of the top 100 in the Lown ranking, 91 are nonprofits. Only nine are safety-net hospitals, even though they account for 21% of the total. That’s probably because these hospitals tend to serve poorer communities, and even if they provide excellent care, patients might not receive the high-quality follow-up care that can keep them alive and healthy. Quality of care includes the percentage of patients who die one month, three months, and one year after their hospital stay.

“Life expectancy,” said Saini, often “depends more on your ZIP code than your genetic code” — an argument for hospitals to contribute to community groups that make a ZIP code healthier.

The top 100 also includes 53 teaching hospitals, which make up only one-third of all hospitals. Their overrepresentation reflects generally excellent patient outcomes and a culture of avoiding procedures that lack scientific evidence of efficacy.

The starkest disparity is between many hospitals’ excellent patient outcomes and their poor civic leadership, as measured by Lown. (It used Internal Revenue Service and other government data for CEO pay and spending on charity care and community organizations.) The leaders for civic leadership were Harris Health System in Houston and five New York City public hospitals.

On average, 2.8% of hospital expenses went to charity care. But while the most charitable spent 15% or more, about 150 spent less than 0.1%. And although some nonprofit hospitals, whose tax-exempt status requires them to provide “community benefits,” spent 20% or more on their community, some spent less than 1%.

A number of hospitals renowned for their quality of care got A+ or A for patient outcomes but a D for civic leadership, including Massachusetts General Hospital, the Cleveland Clinic, the Mayo Clinic, and the University of Washington in Seattle. Teaching hospitals such as these do pretty well on supporting their communities, but pay equity — multimillion-dollar salaries for top executives — sinks them in Lown’s calculation of civic leadership. Not a single hospital in the top 100 for patient outcomes was in the top 100 for civic leadership, and vice versa.

Few patients are likely to reject a hospital based on how much it pays its CEO relative to its custodians. But Binder called civic leadership an important metric. “If there is anything you want out of a hospital, it’s ethics,” she said. Charity care, community support, and pay equity “are informative about a hospital’s culture.”